External Email - Use Caution        

Dear Douglas,


Thanks for your advice. I have tried the command that you've suggested. But I still encounter the issue. I am attaching the recon-all.log to this email. The good news is that the other cases were good after using the mri_convert command, except for this subject. 


Dear Bruce,


Yes, the orig.mgz shows the correct orientation in freeview - anatomical location is as expected. Below is the caption of the brain that I'm processing:




Moreover, may I know if it is normal to have lower quality of images in this file?


Thank you so much.


Best regards,

Krystal

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Douglas N. Greve <dgreve@MGH.HARVARD.EDU>
Sent: 20 June 2023 22:14:10
To: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] [FreeSurfer] Talairach Failure Detection
 

        External Email - Use Caution        

It looks like it is failing at the mri_em_register stage, not talairach (I know, confusing). If I had to guess, I would guess that the problem stems from having so much neck/shoulders/back in the image. Can you try an experiment? Try running
mri_convert --slice-crop 100 511 001.mgz 001.cropped.mgz
Look at 001.cropped.mgz to see if it removed a lot/most of the neck, etc. You can try increasing/descreasing the 100 value to get more/less of the neck.  If this works, then try passing 001.cropped.mgz recon-all.

On 6/20/2023 9:21 AM, Fischl, Bruce R.,PHD wrote:

        External Email - Use Caution        

Hmm. Does the orig.mgz show correctly if you just bring it up in freeview (as opposed to using tkregisterfv)? Also, can you send us the recon-all.log?

 

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> On Behalf Of Krystal Xiwing Yau
Sent: Monday, June 19, 2023 10:20 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] [FreeSurfer] Talairach Failure Detection

 

        External Email - Use Caution        

Dear Bruce,

 

Thank you so much for your reply. 

 

The original image is a T1 image in nifti format. 

 

Yes, 001.mgz image looks fine. It shows the correct view whichever I selected:

 

The images are processed in a server by batches - there are 1-2 images have error, while the rest are fine in the same batch. Please kindly advise.

 

Thank you very much again for your help.

 

Best regards,

Krystal

 

 

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Fischl, Bruce R.,PHD <BFISCHL@MGH.HARVARD.EDU>
Sent: 19 June 2023 21:48:08
To: Freesurfer support list
Subject: Re: [Freesurfer] [FreeSurfer] Talairach Failure Detection

 

        External Email - Use Caution        

What was the format of your original image? This generally means that the direction cosines were incorrect in it. Don’t worry about the resampling artifacts – that is because it is interpolating the 001.mgz into the voxel space of the orig.mgz. If you load the 001.mgz without the orig it will look fine.  Are you sure that the 001.mgz looks fine in all views (like are inferior/superior correct)?

 

Cheers

Bruce

 

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> On Behalf Of Krystal Xiwing Yau
Sent: Monday, June 19, 2023 3:51 AM
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] [FreeSurfer] Talairach Failure Detection

 

        External Email - Use Caution        

Dear Freesurfer experts,

 

I would like to seek for help regarding talairach error. I have browsed through the forum about it but still not solving my problem...

 

The following steps are the solutions I have tried:

 

  • Re-run recon-all with -notal-check
  • Re-run recon-all with -notalairach
    • The above flags gave the same error message as follow:
    • WARNING2: gca.c::GCAhistoScaleImageIntensities: h_mri->nbins=100, mri_peak=108
      after smoothing, mri peak at 0, scaling input intensities by inf
      error: Numerical result out of range
      error: GCAhistoScaleImageIntensities: could not find wm peak
      Command exited with non-zero status 252
  • Ran the command to fix the registration: tkregisterfv --s Sub-108 --fstal --no-surfs
    • The view for this subject is weird because it shows the axial view when the coronal view is selected
    • The correct view showed up when the coronal view is selected

 

 

When I viewed 001.mgz in Freeview, there was no such problem. However, some artifact (i.e. the weird lines) occurred in certain slices, as circled in pink:

 

 

I am not sure how to fix it with tkregisterfv and what are the standards for it.

 

It would be so much appreciated if someone could help out with this issue. 

 

Regards,

Krystal

 

Tel.: +852 2255 5024

Fax: +852 2872 5828

Address: Room 304, 3/F, New Clinical Building, 102 Pokfulam Road, Hong Kong.


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer